|Publication number||US7764454 B2|
|Application number||US 11/632,335|
|Publication date||Jul 27, 2010|
|Filing date||Jul 13, 2005|
|Priority date||Jul 13, 2004|
|Also published as||CN1997515A, CN1997515B, EP1768840A2, EP1768840A4, EP1768840B1, US20080084627, WO2006017367A2, WO2006017367A3, WO2006017367B1|
|Publication number||11632335, 632335, PCT/2005/25129, PCT/US/2005/025129, PCT/US/2005/25129, PCT/US/5/025129, PCT/US/5/25129, PCT/US2005/025129, PCT/US2005/25129, PCT/US2005025129, PCT/US200525129, PCT/US5/025129, PCT/US5/25129, PCT/US5025129, PCT/US525129, US 7764454 B2, US 7764454B2, US-B2-7764454, US7764454 B2, US7764454B2|
|Inventors||Igor V. Roshchin, Oleg Petracic, Rafael Morales, Zhi-Pan Li, Xavier Batlle, Ivan K. Schuller|
|Original Assignee||The Regents Of The University Of California|
|Export Citation||BiBTeX, EndNote, RefMan|
|Patent Citations (20), Non-Patent Citations (36), Referenced by (7), Classifications (23), Legal Events (6)|
|External Links: USPTO, USPTO Assignment, Espacenet|
This application claims the benefit of International Patent Application No. PCT/US2005/025129, filed Jul. 13, 2005, which further claims the benefit of U.S. Provisional Patent Application No. 60/587,789 entitled “EXCHANGE-BIAS BASED MULTI-STATE MAGNETIC MEMORY AND LOGIC DEVICES AND MAGNETICALLY STABILIZED MAGNETIC STORAGE” and filed Jul. 13, 2004. Both applications are incorporated herein by reference in their entirety.
This invention was made in part with U.S. Government support under FA9550-04-1-0160 and DE-FG03-87ER-45332. The Government has certain rights in the invention.
This application relates to magnetic materials and magnetic storage media.
Magnetic materials may be used for data storage. Magnetic cards, tapes, disks, and hard drives are examples of storage devices made from various magnetic materials. In most commercial magnetic recording media, each recorded bit has two logic states “0” and “1” that are represented by two different magnetic states of the recording media. Two opposite directions of the magnetization of a magnetic domain in a magnetic material, for example, may be used for the binary states. In such a material, the density of stored information coincides with the density of the magnetic domains or “bits” carrying the information. Therefore, the size of the magnetic domains in such binary magnetic recording media dictates the storage density.
One approach to increasing the magnetic storage density is to reduce the magnetic domain size or the bit size. Many techniques may be used to achieve. For example, high-density recording materials with small domain sizes developed by Fujitsu and other companies have demonstrated storage densities at about 100 Gbit/sq. inch. A decrease in the magnetic bit size, however, may lead to a decrease in the thermal and magnetic stability of the magnetically stored information. Instability in magnetic media may be caused by various factors, including an increase in temperature and by presence of stray fields from the recording head when operating on the adjacent bits, and from the interaction with the adjacent bits.
This application discloses techniques, materials, and devices to achieve high-density magnetic storage based on the magnetic exchange bias caused by magnetic coupling between a ferromagnetic layer and an adjacent antiferromagnetic layer. This magnetic coupling may be used to stabilize a binary magnetic memory device and to construct multi-state magnetic memory devices.
In one implementation, a method is described to stabilize a binary magnetic memory media. This method provides a magnetic storage material which comprises a ferromagnetic layer and an adjacent antiferromagnetic layer that are magnetically coupled to each other to provide a magnetic hysteresis loop as a binary magnetic storage medium. The ferromagnetic layer and the adjacent antiferromagnetic layer are configured to set the center of the magnetic hysteresis loop away from H=0 to stabilize each binary state against a stray magnetic field.
In another implementation, a device is described to include a magnetic storage material which comprises a ferromagnetic layer and an adjacent antiferromagnetic layer that are magnetically coupled to each other to store data at an operating temperature which is lower than a blocking temperature of the antiferromagnetic layer. The blocking temperature is lower than a Curie temperature of the ferromagnetic material. In one configuration of this device, the magnetic material is configured to have more than two magnetic states. In another configuration of this device, the ferromagnetic layer and the antiferromagnetic layer are selected to offset a magnetic hysteresis loop of the material away from a zero field to a magnetic field greater than a stray magnetic field in the device.
The above device may also include a temperature control mechanism to locally control a temperature of a selected magnetic domain and the material when data is written into the selected magnetic domain. The temperature control mechanism may be configured in various configurations. For example, it may include a laser and focuses a laser beam from the laser to a selected magnetic domain during writing, a conducting scanning probe microcopy (SPM) tip which heats up a selected magnetic domain, a resistive (ohmic) heating mechanism as part of each magnetic domain, or a resistive (ohmic) heating element which is in contact with the selected magnetic domain.
This application also describes a method for magnetic recording. A magnetic storage material is provided to include a ferromagnetic layer and an adjacent antiferromagnetic layer that are magnetically coupled to each other to store data at an operating temperature which is lower than a blocking temperature of the antiferromagnetic layer. The blocking temperature is lower than a Curie temperature of the ferromagnetic material. Localized heating is performed to raise temperature of a selected magnetic bit in the material above the blocking temperature. At the temperature above the blocking temperature, a magnetization process is applied to the selected magnetic bit set a remanent magnetization to a selected value of a pre-determined set of remanent magnetizations for the multi-state storage medium to write a bit. The external magnetic field at the selected magnetic bit is then removed and, at the zero field, the temperature of the selected magnetic bit is reduced below the blocking temperature to store the remanent magnetization in the selected bit.
This application further describes another method for magnetic recording. A magnetic storage material is provided to include a ferromagnetic layer and an adjacent antiferromagnetic layer that are magnetically coupled to each other to store data at an operating temperature which is lower than a blocking temperature of the antiferromagnetic layer. The blocking temperature is lower than a Curie temperature of the ferromagnetic material. Localized heating is performed to raise temperature of a selected magnetic bit in the material above the blocking temperature. At the temperature above the blocking temperature, a magnetization process is applied to the selected magnetic bit to set a remanent magnetization to a selected value of a pre-determined set of remanent magnetizations for the multi-state storage medium to write a bit. While maintaining a magnetic field of the magnetization process, the temperature of the selected magnetic bit is reduced below the blocking temperature to store the remanent magnetization in the selected bit.
These and other implementations, their application and variations are described in greater detail in the attached drawings, the detailed description, and the claims.
The magnetic materials, storage techniques and storage devices described in this application use two magnetically coupled magnetic layers, i.e., a ferromagnetic (F or FM) layer and an adjacent antiferromagnetic (AF or AFM) layer, to provide stabilized binary magnetic storage media and multi-state magnetic storage media. Stabilized binary magnetic storage media described here are tolerant to magnetic interference caused by various factors, including stray magnetic fields from the read/write head, adjacent magnetic bits, and other sources and hence can be used to achieve highly dense magnetic bits on the media. Multi-state magnetic storage media described here provide multiple remanent magnetization values to represent multiple states in each magnetic bit to increase the information content stored in each bit.
The local heating may be implemented in various configurations. For example, the temperature control 220 may use a laser to direct a focused laser beam to a selected bit on the medium 201 for the local heating. The focusing of the laser beam may be achieved by using near-field optics such as a solid immersion lens. In another example, electrical resistive (ohmic) heating may be used to control the heating and the temperature of each individual bit position in the F-AF media. The resistive heating may be implemented as part of the F-AF media to heat each bit itself by running a heating current at each bit position. Alternatively, a separate resistive heating element may be used to be in contact with the area of the bit, e.g. at the intersection of the bit/word lines like in the writing of a bit in a CMOS-type memory. In yet another example, a conducting scanning probe microcopy (SPM) tip can be used to direct a current to a selected bit for local heating. Notably, the thermal control 220 may be integrated as part of the read/write head 210 so that a single positioning mechanism can be used to control the positioning of the local heating and the magnetic head.
The F-AF materials described in this application may also be used to construct MRAM without any moving parts. Such MRAM includes a temperature control to individually control the temperature at each bit position of the MRAM.
The following sections first describe materials designs and properties of F-AF coupled media based on exchange bias. Next, domain stabilization for binary F-AF media and multi-state F-AF media are described.
Properties of Ferromagnetic-Antiferromagnetic Coupled Media
Exchange bias due to magnetic coupling of F and AF layers causes a change in the magnetic anisotropy in both F and AF layers. One theoretical model for explaining the exchange bias is the lateral length scale model. See, Igor Roshschin et al. in “Lateral length scales in exchange bias,” Europhysics Letters published on line at http:/www.edpsciences.org/epl on Jun. 17, 2005, and is scheduled to appear in Vol. 71 (July 2005), the entire disclosure of which is incorporated by reference as part of the specification of this application. Other explanations are possible. Experimental findings explaining the local regime of exchange bias are used here to simply aid the understanding of the exchange bias behind the stabilization techniques and multi-state memory techniques and should be construed to limit various implementations described in this application.
When a ferromagnet is in proximity to an antiferromagnet, lateral length scales such as the respective magnetic domain sizes can drastically affect the exchange bias. Bilayers of FeF2 and either Ni, Co or Fe are studied using SQUID and spatially resolved MOKE. When the antiferromagnetic domains are larger than or comparable to the ferromagnetic domains, a local, non-averaging exchange bias is observed. This gives rise to unusual and tunable magnetic hysteresis curves.
When two dissimilar materials are in contact, the proximity effect is often observed, where one material modifies the properties of the other. Because of the finite extent of electron wave functions, the proximity effect is typically described as the spatial variation of an order parameter across the interface. For example, the proximity effect that occurs in superconductor-normal bilayers is characterized by the decay of the superconducting order parameter into the normal material with a length scale referred to as the coherence length. Interestingly, no correlation between characteristic length scales of physical quantities changing in the plane parallel to the interface has ever been claimed to play a role in the proximity effect.
The relation between the lateral characteristic length scales on the two sides of the interface is important for exchange bias. Exchange bias (EB) is a proximity effect between a ferromagnet (F) and an antiferromagnet (AF) in intimate contact with each other . Usually, EB is described as an additional unidirectional anisotropy induced by the AF into the F via exchange coupling at the interface. This produces a single magnetic hysteresis loop shifted along the magnetic field axis, below the AF ordering (Néel) temperature, TN. The magnitude of this shift is defined as the exchange bias field, He, and can be positive or negative. Unlike AF affecting F, less obvious and usually more difficult to observe is the effect of the F on the properties of the AF . Both the F and AF can form magnetic domains, which leads to spatial variation of the magnetization in the F and the staggered magnetization in the AF in the plane parallel to the interface. Due to the mutual influence between the F and AF, the interfacial spin configuration may deviate significantly from that of the bulk AF and F. In this report, the AF domains in the exchange-biased systems refer to the area that induces the same unidirectional anisotropy.
When the AF domains are larger than or comparable to the F domains, local, non-averaging EB is observed . Such a sample can split magnetically into two subsystems with the He of the same magnitude but opposite sign. This state is achieved either by zero-field cooling a partially demagnetized sample or by cooling the sample in a properly chosen constant applied magnetic field. In either case, the two magnetic subsystems behave independently of each other, and no averaging of EB occurs. This manifests itself clearly as a double hysteresis loop. In addition, the local EB sign can be manipulated by either creating domains in the F during the demagnetization process or by varying the field-cooling procedure . These results reveal new physics of heterogeneous magnetic structures and proximity effects with inhomogeneous order parameters.
Typically, a 38-100 nm thick layer of FeF2 is grown on a (110) MgF2 substrate at 300° C., followed by a 4-70 nm thick ferromagnetic layer (Co, Fe or Ni) grown at 150° C., coated in situ with a 3-4 nm layer of Al to prevent the magnetic layers from oxidation . The FeF2 antiferromagnet (TN=78.4 K) grows epitaxially and untwinned in the (110) orientation on a (110) MgF2 substrate  as determined by X-ray diffraction. Based on the bulk structure, the ideal surface of (110) FeF2 is assumed to have compensated spins, oriented in-plane with the easy axis along the  direction . X-ray diffraction measurements show that the ferromagnetic layer for all samples is polycrystalline. A uniaxial magnetic anisotropy present in all samples, with the easy axis along the FeF2 easy axis, even at temperatures much higher than TN is attributed to a growth-induced anisotropy. The ferromagnetic transition temperature of all three ferromagnets is well above room temperature. The in-plane sample magnetization parallel to the applied field is measured using a dc SQUID magnetometer and magneto-optical Kerr effect (MOKE). The direction of the sample magnetization during cool down is defined as positive; the sign of the hysteresis loop shift defines the EB sign. The magnetic moment vs. applied magnetic field curves for all samples show a typical ferromagnetic behavior above TN: a single hysteresis loop, symmetric with respect to the origin.
In the first series of experiments, the sample is demagnetized at 300 K to a chosen value of the remanent magnetization along the easy axis, MR, between 0 and MS, the saturation magnetization. This leads to the formation of F domains with the magnetizations in opposite directions along the easy axis. The balance between the magnetization of the two types of domains determines the resultant magnetization of the sample. After the sample is cooled in zero magnetic field (ZFC) below TN, the low-field magnetic moment is measured as a function of applied magnetic field at various temperatures. The sample cooled from full remanent magnetization, MR≈MS, shows single hysteresis loops exchange-biased to negative fields by He(T) (
In the second series of experiments, the same sample is first magnetized at T=150 K by applying the magnetic field along the easy axis, above the irreversibility field of the hysteresis curve. Then the sample is cooled in an applied field (FC), HFC (
It is remarkable that in both experimental series, positive and negative He(T) of equal absolute value are found at all temperatures below TN (
In these samples, the loop half-width is smaller than its shift, i.e. Hc<He. The condition Hc≦He is essential for clear observation of the double hysteresis loops. In contrast, the results presented in [7,8] were in the opposite limit: Hc>>He, thus the double loops could not be resolved.
The conclusions of the first two series of experiments are unambiguously confirmed by spatially resolved MOKE, measured as intensity difference of the reflected p-polarized light .
For this experiment, a 5 mm×5 mm FeF2 (70 nm)/Ni(70 nm)/Al (4 nm) sample is demagnetized above TN and then cooled below TN in a zero applied magnetic field as is done in the first series of the experiments. First, the MOKE signal (generally considered  to be proportional to the magnetization) as a function of applied magnetic field is collected from the entire sample surface area, illuminated with a wide beam. The curve consists of a double hysteresis loop (
This experiment confirms that the sample has two types of areas; single hysteresis loops are shifted by the same He, positively in one area, and negatively in the other. When the laser beam covers parts of both areas, the magnetization curve consists of two loops shifted in opposite directions, with their relative heights determined by the ratio of the two areas.
These experimental results lead to two main findings. First, the sample can split into two independent subsystems with opposite sign of the EB. This can be achieved either by cooling a partially or completely demagnetized sample in zero applied magnetic field, or by cooling the sample in a judiciously chosen intermediate applied magnetic field. The appropriate field range is determined by sample parameters: materials, layer thickness, interface roughness, etc. Second, the exchange bias for each F domain is not averaged over various AF domains, so double hysteresis loops are observed.
The phenomenon of EB has attracted much attention; several mechanisms of EB have been proposed of which many neglect variations of EB on the scale of F domains. The dependence of Hc and He on the AF and/or F domain sizes in patterned F-AF bilayers, or in diluted AF has been studied. Enhancement of He in the AF-F bilayers with patterned F layer was explained by suppression of F-F exchange interaction in the small F domains. Correspondence of AF domains to the F domains, and small spatial variations of the value of EB were reported. Double hysteresis loops and a correlation of domain structure in the F and the AF were observed in CoO/NiFe samples zero-field-cooled from a demagnetized state and explained using the exchange spring model [20, 21]. Exchange bias field and the shape of the double hysteresis loops in FeF2/Fe  and Fe0.6Zn0.4F2/Fe  bilayers were found to depend on the remanent magnetization during cooling in a zero applied magnetic field. Double and negatively shifted single hysteresis loops were observed in FeMn/FeMnC samples  deposited in various small magnetic fields (0-40 Oe) without any clear trend. In zero-field deposited NiFe/NiFeMn samples, spatial variation of MOKE signal measured along the hard axis was claimed , but no supporting plot was presented.
The large number of reported experiments performed on different samples, show seemingly contradictory and unconnected results. Combined with multiple interpretations, this leads to a complex and confusing situation where no conclusion can be made. It is obvious that only a comprehensive study with several different experiments on the same sample can provide definite answers on the role of the relative sizes of the AF and F domains in EB.
In EB, we find two distinct regimes depending on the relation between the AF and F domain sizes. First, when the AF domains are smaller than the F domains, a F domain averages the direction and magnitude of He over several AF domains. This may lead to a continuous variation of He as a function of the cooling field and single hysteresis loops, as found in twinned FeF2/Fe  and MnF2/Fe  samples. If the AF domains are very small, the number of the AF domains interacting with each F domain is very large. In this case, each F domain can be essentially treated as a separate sample. When ZFC the sample, the sign for the net EB for each of these domains is set by the direction of the magnetization of the F domain, so double loops may occasionally be observed [20-22]. X-ray reflectivity measurements  imply that the in-plane structural coherence in the twinned (110) FeF2 grown on (100) MgO is 6-10 nm with the easy axes of the crystallites at 90° to each other. Such spatial variation of the easy axis, combined with the small size of the grain establishes the maximum size of AF domains, resulting in the averaging regime of EB.
When the AF domain size is comparable to or larger than the F domain size, each F domain couples only to one AF domain with a particular direction of the EB, so no averaging occurs. This results in the same absolute value of the He for different cooling fields and double hysteresis loops, as reported in this work. In this regime, when the sample is zero-fieldcooled, the unidirectional anisotropy in the AF is set locally through the antiferromagnetic coupling  between the interfacial F and AF moments. Therefore, the local sign of the EB for each F domain is determined by the direction of the magnetization of that domain during sample cooling. Thus, domains of the F with opposite magnetization direction above TN have the same magnitude but opposite sign of EB below TN.
When the sample is field-cooled, also, only the sign, but not the magnitude of He depends on the cooling field, HFC. It is the sign of the projection of Htotal on the easy axis, Htotal, that determines the sign of the local EB. Htotal, the total local cooling field sensed by the interfacial spins of the AF, is the sum of the applied magnetic field, HFC, and the local exchange field due to the interfacial F moments, HF. Spatial inhomogeneity of the sample leads to inhomogeneity of HF, which in turn leads to inhomogeneity of Htotal. At intermediate HFC, this causes the magnitude of Htotal to be small and spatially inhomogeneous, giving rise to areas of the interface with positive and negative sign of Htotal. Consequently, cooling the sample below TN in this field leads to the negative and the positive EB, respectively. According to X-ray reflectivity measurements  the in-plane structural coherence in the untwinned (110) FeF2 grown on a (110) MgF2 single crystal substrate is ˜28 nm. Moreover, due to absence of twinning, the size of AF domains can be much larger than the grain size, unlike in the case of a twinned FeF2. Large, mm-size AF domains have been observed in bulk single-crystal fluorides . In the ZFC case, the AF and F domains are observed to be comparable in size and larger than the laser beam. Simulations confirm that the relative domain sizes determine the splitting of the sample into two magnetic subsystems and the regime of EB (to be published elsewhere).
The relevance of the observation presented here to other systems was illustrated in a multilayer system imitating an AF-F system . For the sample with the larger domains in the layer playing the role of interfacial AF spins, double hysteresis loops were observed for intermediate cooling fields. It is not clear whether such a model system correctly mimics the domain structure in real AF-F systems. Those results, however, provide an additional experimental configuration in which relative domain sizes play a role.
Hence, the relative AF and F domain sizes affect ex-change bias. When the AF domain size is larger than or comparable to that in the F, structural or magnetic inhomogeneities can result in the sample splitting into two independent subsystems upon cooling through the AF transition temperature, TN. Each subsystem exhibits EB of the same magnitude but of the opposite sign. This results in magnetization curves with double hysteresis loops, thus unambiguously showing non-averaging, local EB. This behavior is reproducibly observed in samples with three different F: Ni, Fe, and Co, with various F and AF layer thicknesses. These observations also show that not only the properties of the F but those of the AF are modified in EB. We suggest that relative lateral length scales may be relevant in other proximity effects with spatially inhomogeneous order parameters, such as a granular superconductor (S) in proximity to a normal metal, or a S in proximity to a F that has domains .
The present magnetic storage materials and devices based on magnetic coupling between a ferromagnetic (F) layer and an adjacent antiferromagnetic (AF) layer use the remanent magnetization in the F layer to imprint the AF layer. This AF-F coupling produces exchange bias.
Magnetically Stabilized Binary Recording Media
The above domain imprinting allows controlling locally the direction of the additional unidirectional anisotropy induced into the F from the AF. This additional unidirectional anisotropy enhances the magnetic stability of the F bit or F nanostructure by increasing the effective energy barrier for magnetic switching.
Hence, the present use of the F-AF coupling in magnetic media allows smaller bit size and alleviates certain limitations in the ferromagnetic storage media. As mentioned above, the ferromagnetic storage medium is adversely affected by the stray field produced by the recording head operating on the adjacent bit due to the danger of partial or full erasure of the stored information. The design of F-AF coupled media to shift the hysteresis loop and to increase the coercive field of the media can significantly reduce the adverse effects of the stray fields. As another example, the present technique can also remove or lessen the restriction on the operating temperature range due to the danger of partial or full erasure of the stored information by thermal fluctuations of the magnetic moments within the magnetic bit. The present magnetic stabilization may be used in magnetic recording such as high-density and high-stability hard disk drives, magnetic tapes, magnetic RAM, and other magnetic memory and logic devices.
The present stabilization technique may be used in magnetic recording devices and magnetic recording media for both in-plane and perpendicular magnetic recording to stabilize the information stored in patterned or unpatterned magnetic media and to increase the recording density and reach the range of 100s Gbit/sq. inch and 1 Tbit/sq. inch densities and higher. In addition, the present stabilization techniques may be used to increase the density of information on magnetic recording tapes and to increase the density and stability of recorded information in Magnetic Random Access Memory (MRAM). Furthermore, the present stabilization techniques may also be used in other magnetic recording applications for similar purposes.
The described stabilization techniques may be applied to magnetic recording media having thin film multilayers that include a ferromagnet and either an antiferromagnet, or a so-called synthetic antiferromagnet (SAF), which is hereafter included into the AF category. In the bit recording process, the recording head heats up locally the bit containing the F and AF layers, above the AF blocking temperature, TB. In case of Magnetic RAM type of memory, heating can be done resistively, using Ohmic heat. The F layer magnetization of the bit is switched by applying a local magnetic field produced by the recording head in a fashion similar to the existing, or any proposed in the future technology. The heating is then switched off, and the bit is cooled below TB, so that the unidirectional anisotropy of the AF layer that gets induced into the F, is set.
After that the magnetic field can be turned off. As an alternative for the process, the magnetic field can be turned off before cooling the sample to achieve the same result as demonstrated in
The readout can be done employing existing, or future magnetic information readout technology. The read head senses the stray magnetic fields due to the magnetic moment of the F layer of the bit. Due to exchange bias with the AF layer, the magnetic hysteresis loop of such bit is shifted to positive or negative fields. If the loop is negatively shifted, the remanent magnetization is positive, and vice-versa. This defines “0” and “1” state. AF materials need to have TB>80° C. This includes, but is not limited to: NiO, IrMn, and FeMn.
The F-AF coupled media may also be patterned to separate different magnetic bits from one another to further improve the stability and the tolerance to stray fields.
Magnetic Storage with Multiple States from AF-F Coupling
The above imprinting from the F layer to the AF layer can be used to locally set the direction of the exchange bias to either positive or negative. Therefore, a bi-domain magnetic state can be created where the two domains have opposite directions of the exchange bias. The ratio of the sizes of the two domains determines the remanent magnetization of the magnetic area which is used as a magnetic “bit” for data storage. This use of two domains at one bit location allows creating a magnetic memory with multi-state magnetic “bits,” which are well-defined and stable in a wide range of magnetic fields. The states correspond to the values of the remanent magnetization, Mr, between the positive and negative values of the saturated magnetization Ms such that Mri=Ms/n, where 2n+1 is the total number of states, and i can take any values between −n and +n. The maximum number of the multi states is set by the magnetic sensing limit of the device.
In such multistate storage devices, the readout procedure of such multi-state “bits” can be done by various methods, including magneto-optical or magneto-resistive methods. The multi-state magnetic materials may be used in multi-state magnetic recording devices and magnetic recording media for both in-plane and perpendicular magnetic recording, and in multi-state logic elements. Among a number of the advantages, such materials may be used to increase the magnetic memory density by recording information in a multi-state magnetic “bit” that stores more than one bit of information. Multi-state magnetic logic elements and devices can be created based on the above multi-state bits.
In various implementations, such materials may be used to form thin film multilayers that include a ferromagnet and an antiferromagnet, or a so-called synthetic antiferromagnet (SAF), hereafter included into the category “AF.” The ratio of the sizes of the two domains determines the remanent magnetization of the magnetic area as a magnetic “bit”). The exact value of remanent magnetization can be set and controlled by various magnetization processes.
Various writing methods may be used to produce the desired remanent magnetization levels in each bit. Two examples are described below.
1. The chosen magnetic bit is heated above its blocking temperature, TB, i.e. the temperature at which the exchange bias direction is set.
2. The chosen remanent magnetization, Mr0 is achieved by following a minor loop. One of the possible procedures is to saturate the magnetization of the magnetic bit by applying a large enough magnetic field in a direction opposite to the direction of Mr0, and then apply a smaller magnetic field in the direction of Mr0 of such a magnitude to produce the remanent magnetization to be equal to Mr0 upon removal of the applied magnetic field. Then the magnetic field is removed.
3. The bit is cooled below TB in absence of an applied magnetic field.
4. The remanent magnetization value is imprinted in this AF-F bit, and is not affected by either the magnetic field used for writing adjacent magnetic bits, nor by the stray fields from the adjacent magnetic bits, as long as the particular bit is kept below TB.
A second writing process is described below where an external field is applied during the cooling process:
1. The chosen magnetic bit is heated above its blocking temperature, TB, i.e. the temperature at which the exchange bias direction is set.
2. The magnetic field of a particular magnitude is applied.
3. Then the bit is cooled below TB in the applied magnetic field.
4. Cooling the bit in a magnetic field of a particular magnitude allows to set a bi-domain structure that has a double magnetic hysteresis loop with a chosen value of the remanent magnetization. The remanent magnetization value is imprinted in this AF-F bit, and is not affected by the magnetic field used for writing adjacent magnetic bits, as well as the stray fields from the adjacent magnetic bits, as long as the particular bit is kept below TB.
The recorded signal can be set to correspond to any chosen remanent magnetization of the bit in the range between −Ms and Ms that is very stable in a wide field range. Local heating can be performed by any method, e.g. using a laser, near-field optics, conducting scanning probe microcopy (SPM) tip or resistive (Ohmic) heating of the bit itself or the element in contact with the area of the bit, using, e.g. design with a combination of the bit/word lines like writing a bit in CMOS-type memory. AF materials need to have TB>80° C. to be useful for applications. Examples of suitable AF materials include, but are not limited to, NiO, IrMn, and FeMn. The readout procedure of this multi-state “bits” can be done by various methods, including magneto-optical or magneto-resistive methods. One of the possible scenarios is to measure the magneto-optical Kerr effect (MOKE) signal that is proportional to the value of the remanent magnetization, which corresponds to the recorded state.
Such multi-state bits may be used as a basis for a new generation of multi-state logic elements and devices. In this case, as an alternative to the laser heating, resistive heating can be used in the writing procedure described above. The readout can be either magneto-optical or magnetoresistive. An example for the possible magnetoresistive readout could be a magnetoresistive sensor on top of the magnetic “bit.” The value of the resistance of that sensor should uniquely correspond (be proportional) to the value of the local magnetic field, similarly to the technique used in the GMR read-heads.
Applications of such and other materials include magnetic recording such as high density and high stability hard disk drives, magnetic tapes, Magnetic RAM, other magnetic recording devices, and multi-state logic devices and elements of logic circuits. Other applications are also possible.
Only a few examples and applications are described. Other variations and enhancements may be made.
|Cited Patent||Filing date||Publication date||Applicant||Title|
|US3521294 *||Mar 13, 1967||Jul 21, 1970||Ampex||Magneto thermal recording process and apparatus|
|US4385273||Sep 18, 1980||May 24, 1983||Lgz Landis & Gyr Zug Ag||Transducer for measuring a current-generated magnetic field|
|US5889641 *||May 5, 1997||Mar 30, 1999||Seagate Technology, Inc.||Magneto-resistive magneto-optical head|
|US6317280 *||Sep 24, 1998||Nov 13, 2001||Sharp Kabushiki Kaisha||Thermomagnetic recording and reproducing head having a heating head with a width narrower than the magnetic heads, a recording and reproducing device with such a heating head and methods related thereto|
|US6327227 *||Jul 24, 1998||Dec 4, 2001||Sharp Kabushiki Kaisha||Magnetic recorder/reproduction apparatus including a selective temperature controller and magnetic recording reproduction method including selective temperature control|
|US6532125 *||Sep 30, 1999||Mar 11, 2003||International Business Machines Corporation||Apparatus and method suitable for magnetic-thermal recording|
|US6777730 *||Aug 29, 2002||Aug 17, 2004||Nve Corporation||Antiparallel magnetoresistive memory cells|
|US6830837 *||Jan 17, 2002||Dec 14, 2004||Hitachi Global Storage Technologies Japan, Ltd.||Magnetic recording medium and magnetic recording apparatus|
|US6834026 *||Jul 23, 2003||Dec 21, 2004||International Business Machines Corporation||Method for thermally-assisted recording on a magnetic recording disk|
|US7502188 *||Sep 24, 2004||Mar 10, 2009||Fujitsu Limited||Magnetic recording medium, magnetic storage and method for reproducing information from magnetic recording medium|
|US20040086750 *||Apr 5, 2002||May 6, 2004||Oliver De Haas||Antiferromagnetic layer system and methods for magnectically storing data in anti-ferromagnetic layer system of the like|
|US20060062132 *||Sep 26, 2003||Mar 23, 2006||Koninklijke Philips Electronics, N.V.||Thermally-assisted recording medium with a storage layer of antiferromagnetic double-layer structure with anti-parallel orientation of magnetization|
|US20060147758 *||Jan 6, 2005||Jul 6, 2006||Hong-Sik Jung||Perpendicular magnetic recording medium with magnetically resetable single domain soft magnetic underlayer|
|US20070109838 *||Oct 12, 2004||May 17, 2007||Agency For Science, Technology And Research||Magnetic memory device|
|JPH01220239A||Title not available|
|JPH08273222A||Title not available|
|JPS61104442A||Title not available|
|JPS62184644A||Title not available|
|WO2002084647A2||Apr 5, 2002||Oct 24, 2002||Leibniz-Institut Für Festkörper- Und Werkstoffforschung Dresden E.V.||Antiferromagnetic layer system and methods for magnetically storing data in antiferromagnetic layer systems of the like|
|WO2006017367A2||Jul 13, 2005||Feb 16, 2006||The Regents Of The University Of California||Exchange-bias based multi-state magnetic memory and logic devices and magnetically stabilized magnetic storage|
|1||Aladyshkin, A. Yu. et al., "Domain-wall superconductivity in hybrid superconductor-ferromagnet structures," Phys. Rev. B 68: 184508-1-184508-7, (2003).|
|2||Berkowitz, A.E. and K. Takano, "Exchange anisotropy-a review," J. Magn. Magn. Mater. 200: 552-570 (1999).|
|3||Berkowitz, A.E. and K. Takano, "Exchange anisotropy—a review," J. Magn. Magn. Mater. 200: 552-570 (1999).|
|4||Chien, C.L., et al., "Hybrid domain walls and antiferromagnetic domains in exchange-coupled ferromagnet/antiferromagnet bilayers," Phys.Rev. B. 68: 014418-1-014418-5 (2003).|
|5||Cowburn, R.P. "Superparamagnetism and the future of magnetic random access memory," J. Appl. Phys. 93(11): 9310-9315 (2003).|
|6||Eisenmenger, J. and I.K. Schuller, "Magnetic Nanostructures: Overcoming thermal fluctuations," Nature Materials 2: 437-438 (Jul. 2003).|
|7||English language translation of Japanese Office Action dated Nov. 19, 2008 for JP Patent Application No. 2007-521668, 7 pages.|
|8||English language translation of Korean Intellectual Property Office, Final Office Action dated Jan. 22, 2009 for Korean Patent Application No. 10/2007-7003212, 3 pages.|
|9||English language translation of Korean Intellectual Property Office, Office Action dated Jan. 30, 2008 for Korean Patent Application No. 10/2007-7003212, 3 pages.|
|10||Fullerton, E.E. et al., "Antiferromagnetically coupled magnetic media layers for thermally stable high-density recording," Appl. Phys. Lett. 77(23): 3806-3808 (2000).|
|11||Gökemeijer, N.J. et al., "Memory effects of exchanging coupling in ferromagnet/antiferromagnet bilayers," Phys. Rev. B. 60(5): 3033-3036 (1999).|
|12||International Search Report and Written Opinion dated Jun. 16, 2006 for PCT/US2005/25129, published as WO 2006/017367, and entitled: "Exchange-Bias Based Multi-State Magnetic Memory and Logic Devices and Magnetically Stabilized Magnetic Storage" (7 pages).|
|13||Japanese Patent Office Action dated Nov. 19, 2008 for JP Patent Application No. 2007-521668,5 pages.|
|14||Judy, J.H., "Past, present, and future of perpendicular magnetic recording," Journal Magnetic and Magnetic Materials 235: 235-240 (2001).|
|15||Kappenberger, P. et al., "Direct imaging and determination of the uncompensated spin density in exchange-biased CoO/(CoPt) multilayers," Physical Review Letters 91(26): 267202-1-267202-4 (2003).|
|16||Kirk, T.L. et al., "Coercivity mechanisms in positive exchange-biased Co films and Co/PT multilayers," Phys. Rev. B 65: 224426-1-224426-5 (2002).|
|17||Kiwi, M., "Exchange bias theory," Journal of Magnetism and Magnetic Materials 234: 584-595 (2001).|
|18||Korean Intellectual Property Office, Final Office Action dated Jan. 22, 2009 for Korean Patent Application No. 10/2007-7003212, 3 pages.|
|19||Korean Intellectual Property Office, Office Action dated Jan. 30, 2008 for Korean Patent Application No. 10/2007-7003212, 5 pages.|
|20||Lange et al., "Phase diagram of a superconductor/ferromagnet bilayer," Phys. Rev. B. 68: 174522-1-174522-6 (2003).|
|21||Liu, K. et al., "Fabrication and thermal stability of arrays of Fe nanodots," Applied Physics Letters vol. 81, No. 2, pp. 4434-4436 Dec. 2, 2002.|
|22||Meiklejohn, W.H. and C.P. Bean, "New magnetic anisotrophy," Phys. Rev. 102(5): 1413-1414 (1956).|
|23||Miltényi, P. et al., "Tuning exchange bias," Appl. Phys. Lett. 75: 2304-2306 (1999).|
|24||Nogués, J. and I.K. Schuller, "Exchange Bias," Journal of Magnetism and Magnetic Materials 192: 203-232 (1999).|
|25||Nogués, J. et al., "Positive Exchange Bias in FeF2-Fe Bilayers," Phys. Rev. Lett. 76(24): 4624-4627 (1996).|
|26||Nogués, J. et al., "Role of interfacial structure on exchange-biased FeFe2-Fe," Phys. Rev. B 59(10): 6984-6993 (1999).|
|27||Nolting, F. et al., "Direct observation of the alignment of ferromagnetic spins by antiferromagnetic spins," Nature 405: 767-769 (2000).|
|28||Ohta, T. et al., "Opitcal recording; phase-change and magneto-optical recording," Journal of Magnetism and Magnetic Materials 242-245: 108-115 (2002).|
|29||Oppeneer, P.M., "Magneto-Optical Kerr Spectra," Chapter 2 in Handbook of Magnetic Materials, vol. 13 (Buschow, K.H. J., Editor), Elsevier, Amsterdam 2001, pp. 229-422.|
|30||Roshchin, I.V. et al., "Magnetic Nanodots: Exchange Bias, Vortex State, and Neutron Studies," American Physical Society, March Meeting 2004, Mar. 22-26, 2004, Palais des Congres de Montreal, Montreal, Quebec, Canada, Meeting ID: MAR04, abstract # H23.013 (Bulletin of the American Physical Society, vol. 49, 2004).|
|31||Roshcin, I.V. et al., "Lateral length scales in exchange bias," Europhysics Letters, vol. 71, No. 2, pp. 297-303 (2005).|
|32||Roy, S. et al., "In-Plan and Depth Dependent Magnetization in FeCr Multilayer," American Physical Society, March Meeting 2004, Mar. 22-26, 2004, Palais des Congres de Montreal, Montreal, Quebec, Canada, Meeting ID: MAR04, abstract # W27 14, Bulletin of the American Physical Society, vol. 49, p. 1334.|
|33||Schuller, I.K. et al., "Imprinting Domain Wall into Antiferromagnet," American Physical Society, March Meeting 2004, Mar. 22-26, 2004, Palais des Congres de Montreal, Montreal, Quebec, Canada, Meeting ID: MAR04, abstract #W27.009 (Bulletin of the American Physical Society, vol. 49, 2004).|
|34||Shi, H. and D. Lederman, "Exchange bias flop in FexZn1-xF2/Co bilayers," Phys. Rev. B 66: 094426-1-094426-6 (2002).|
|35||Stamps, R.L., "Mechanisms for exchange bias," J. Phys. D: Appl. Phys. 33: R247-R268 (2000).|
|36||Wang, J.P., "Design of laminated antiferromagnetically coupled media for beyond 100 Gb/in2 areal density," Journal of Applied Physics 91: 7694-7696 (2002).|
|Citing Patent||Filing date||Publication date||Applicant||Title|
|US8467234 *||Feb 8, 2011||Jun 18, 2013||Crocus Technology Inc.||Magnetic random access memory devices configured for self-referenced read operation|
|US8488372||Jun 10, 2011||Jul 16, 2013||Crocus Technology Inc.||Magnetic random access memory devices including multi-bit cells|
|US8576615||Jun 10, 2011||Nov 5, 2013||Crocus Technology Inc.||Magnetic random access memory devices including multi-bit cells|
|US8854871 *||Nov 18, 2013||Oct 7, 2014||U.S. Department Of Energy||Dynamic control of spin states in interacting magnetic elements|
|US9001630 *||Apr 24, 2014||Apr 7, 2015||Western Digital Technologies, Inc.||Energy assisted magnetic recording medium capable of suppressing high DC readback noise|
|US20120201074 *||Feb 8, 2011||Aug 9, 2012||Neal Berger||Magnetic Random Access Memory Devices Configured for Self-Referenced Read Operation|
|US20150063019 *||Apr 9, 2013||Mar 5, 2015||Commissariat A L'energie Atomique Et Aux Energies Alternatives||Thermally-assisted magnetic writing device|
|Cooperative Classification||B82Y10/00, G11B2005/0002, G11B11/10586, G11B5/66, G11C11/16, G11B11/10528, G11C11/5607, G11B11/10504, G11B11/10584, G11B9/14, G11B5/02, G11B2005/0021|
|European Classification||B82Y10/00, G11B11/105M1, G11B5/66, G11B11/105B6, G11B11/105B1, G11B11/105M2, G11C11/56B, G11C11/16, G11B5/02|
|Sep 4, 2007||AS||Assignment|
Owner name: REGENTS OF THE UNIVERSITY OF CALIFORNIA, THE, CALI
Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:ROSHCHIN, IGOR V.;PETRACIC, OLEG;MORALES, RAFAEL;AND OTHERS;REEL/FRAME:019778/0975;SIGNING DATES FROM 20070105 TO 20070110
Owner name: REGENTS OF THE UNIVERSITY OF CALIFORNIA, THE, CALI
Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:ROSHCHIN, IGOR V.;PETRACIC, OLEG;MORALES, RAFAEL;AND OTHERS;SIGNING DATES FROM 20070105 TO 20070110;REEL/FRAME:019778/0975
|Oct 11, 2007||AS||Assignment|
Owner name: VENCORE SOLUTIONS LLC A DELAWARE LLC, OREGON
Free format text: SECURITY INTEREST;ASSIGNOR:MIGO SOFTWARE, INC. A DELAWARE CORPORATION;REEL/FRAME:019959/0081
Effective date: 20070810
|Mar 23, 2009||AS||Assignment|
Owner name: UNITED STATES AIR FORCE, VIRGINIA
Free format text: CONFIRMATORY LICENSE;ASSIGNOR:CALIFORNIA, UNIVERSITY OF;REEL/FRAME:022438/0957
Effective date: 20080331
|Nov 1, 2010||AS||Assignment|
Owner name: AIR FORCE, UNITED STATES, VIRGINIA
Free format text: CONFIRMATORY LICENSE;ASSIGNOR:CALIFORNIA SAN DIEGO, UNIVERSITY OF;REEL/FRAME:025227/0335
Effective date: 20101008
|Apr 26, 2011||CC||Certificate of correction|
|Jan 27, 2014||FPAY||Fee payment|
Year of fee payment: 4